Download Bacterial pathogens isolated from cultured bullfrogs (Rana

Brief Communications
431
J Vet Diagn Invest 14:431–433 (2002)
Bacterial pathogens isolated from cultured bullfrogs (Rana castesbeiana)
Michael J. Mauel, Debra L. Miller, Kendall S. Frazier, Murray E. Hines II
Abstract. A commercial bullfrog (Rana castesbeiana) operation in south Georgia had multiple epizootics
of systemic bacterial infections over a 3-year period, 1998–2000. A number of potential pathogens (Aeromonas
hydrophila, Chryseobacterium (Flavobacterium) meningosepticum, Chryseobacterium (Flavobacterium) indolgenes, Edwardsiella tarda, Citrobacter freundii, Pseudomonas spp., and (Streptococcus iniae) were isolated
from various tissues. Clinically, frogs demonstrated acute onset of torticolis, stupor, and indifference to stimuli.
Cutaneous hyperemia, subcutaneous and muscular hemorrhage, and peripheral edema were consistent gross
findings. Histologically, clusters of lymphocytes, monocytes, and occasional acidophiles with scattered granulomas occurred in liver, kidney, and spleen. This is the first report of S. inae and C. meningosepticum as
potential disease agents in R. castesbeiana. These findings suggest that a variety of bacteria may be associated
with redleg and that culture results must be obtained for accurate diagnosis.
Domestic farming of frogs is a growing industry, as the
demand for frogs increases for use as pets, food, experimental animals, and educational tools. Farming operations require that frogs be placed in confinement and, as with many
other species, this often leads to an increased risk of disease
and mass mortality.4 Within the aquatic environment, frogs
are in intimate contact with a number of potentially pathogenic bacteria. Under normal conditions, the animals remain
clinically healthy but when stressed by crowding or unsanitary conditions, bacterial opportunists may overcome weakened immune barriers and cause disease. Previous clinical
reports of diseased frogs have implicated Aeromonas hydrophila,3 Citrobacter freundii,3 Acinetobacter lwoffii,4 Flavobacterium spp.,4,7 Pseudomonas spp.,4 Staphylococcus epidermidis,3 Edwardsiella tarda,5 Proteus spp.,4 and Alcaligenes faecalis6 as potential pathogens.
A commercial frog operation in southern Georgia experienced sporadic epizootics of bacterial sepsis (redleg) between 1998 and 2000. At the facility, juvenile frogs are
maintained in aquaria with a continuous (once through) water spray system. As young adults, frogs are transferred to
outdoor ponds and maintained until transferred to surrounding ponds or sold commercially. Frogs in both aquaria and
ponds are kept at very high animal densities that vary with
season and commercial demand. The frogs are fed bulk pelleted ration produced on site.
A total of 17 representative live frogs were submitted for
necropsy on 6 separate occasions. Upon submission for necropsy, frogs were given a routine physical examination and
then euthanized by transdermal exposure to 70% benzocaine
followed by intraperitoneal 50% pentothal. A routine necropsy was performed and representative tissues were submitted for histopathology. Tissues were fixed in 10% buffered formalin, embedded in paraffin, processed routinely,
and stained by hematoxylin and eosin, Gram, and Kinyoun’s
acid-fast methods.
From the Veterinary Diagnostic and Investigational Laboratory,
College of Veterinary Medicine, University of Georgia, PO Box
1389, Tifton, GA 31793.
Received for publication September 25, 2001.
Sections of individual tissues (lung, kidney, spleen, brain,
intestine, and stomach) and an abdominal swab were aseptically collected at necropsy and inoculated onto blood agar
plates and into thioglycollate broth. Inoculated media were
incubated at 30 C with duplicate blood agar plates incubated
in the presence or absence of 5% CO2. Intestine also was
inoculated onto Hektoen enteric agar and incubated at 30 C.
Bacteria selected from pure cultures were stained by the
Gram method, and then the cultures were inoculated according to manufacturers’ instructions into Sensititrea Gram negative AP80 or Gram positive AP90 autoidentification plates,
and the antibiotic sensitivity plate CMVIECOF and allowed
to incubate for 18 hours at 37 C before automated reading
of the reactions. Any isolates that failed to be identified by
the Sensititre system were identified by either the RapID NF
Plus Systemb or the API20E system.c Bacterial isolates, with
antibiotic patterns, from these cases are listed in Table 1.
The Sensititre susceptibility system uses a microversion of
the classic broth dilution method. The NCCLS approved
standard (M31-A) was used to determine the breakpoint concentration of an antimicrobic that inhibits the growth of bacteria. No antimicrobics have been approved for frogs, so
antimicrobics approved for food animals were used. An antimicrobic is listed in Table 1 only if 90% or greater of the
isolates of a particular bacterial species were sensitive or
resistant.
Clinically, the frogs demonstrated acute onset of torticolis,
stupor, and indifference to stimuli. Gross lesions were relatively consistent in frogs from multiple epizootics. Cutaneous hyperemia was noted especially on the extremities and
flanks. Legs were often swollen with marked subcutaneous
edema and focal areas of hemorrhage within the skeletal
muscle. The liver, spleen, and kidney were often severely
enlarged. In a few cases, the liver was discolored with multiple pale foci and the kidneys were usually congested.
Histologically, moderate macrovesicular hepatocellular
vacuolar change, mild cord atrophy, and foci of necrosis
were noted in the liver. Increased numbers of monocytes and
lymphocytes frequently were noted in the liver and kidneys,
with multiple granulomas present in several frogs. All granulomas were negative for acid-fast organisms when using
432
Brief Communications
Table 1. Pathogenic bacteria isolated from farm-raised bullfrogs (Rana catesbeiana) from 1998 through 2000 in southern Georgia and
associated antibiotic sensitivity patterns. An antimicrobic is listed only if 90% or more of the isolates of a particular bacterial species were
sensitive or resistant.
Bacterium
Number of frogs
with isolate
(% of isolates)
Antimicrobic*
Sensitive
Aeromonas hydrophila
7 (18.4)
AP, CF, EN, GM, N
Streptococcus iniae
6 (15.7)
Chryseobacterium meningosepticum
5 (13.2)
AC, AP, CF, CC, EN, E, GM, TP,
N, OX, P, SL, TC, TA, TL, TY
None
Edwardsiella tarda
Citrobacter freundii
5 (13.2)
3 (7.9)
AC, AP, CF, EN, GM, N, SP, TC
AP, CF, EN, GM, N
Chryseobacterium indolgenes
3 (7.9)
EN
Aeromonas spp.
Pseudomonas spp.
5 (13.2)
4 (10.5)
None
None
Resistant
AC, CC, E, OX, P, SP, SL, SU, TI,
TA, TL, TY
SP, SU
AC, AP, CF, CC, EN, E, GM, N,
OX, P, SP, SC, SU, TA, TL, TY
CC, E, P, SU
AC, CC, E, OX, P, SP, SL, SU,
TA, TL, TY
AC, AP, CF, CP, CC, E, GM, P, SP,
SU, TI
None
None
* AC ⫽ ampicillin, AP ⫽ apramycin, CF ⫽ ceftiofur, CC ⫽ clindamycin, EN, enrofloxacin, E ⫽ erythromycin, GM ⫽ gentamycin, N
⫽ neomycin, OX ⫽ oxytetracycline, P ⫽ penicillin, SP ⫽ Spectinomycin, SL ⫽ sulfachloropyridazine, SU ⫽ sulfadimethoxine, TE ⫽
tetracycline, TA ⫽ tiamulin, TL ⫽ tilmicosin, TY ⫽ tylosin.
acid-fast stains, although small rod-shaped bacteria were occasionally identified in these tissues stained by the hematoxylin and eosin method. The skin contained multifocal areas
of epidermal degeneration with abundant dermal edema and
a mixed dermal inflammatory infiltrate that varied from mild
to severe. In a few cases, the epidermis was eroded or focally
ulcerated, with bacteria present in subepidermal vesicles.
Blood vessels in the skin and multiple organs were congested and hyperemic. The skeletal muscle in most of the
frogs examined contained moderate multifocal hemorrhage
and myocyte necrosis (characterized by myofragmentation,
loss of striations, hypereosinophilia, loss of nuclei, and karyorrhectic debris), perimyseal edema, and mild accumulations
of acidophils and monocytes. Bacterial rods were commonly
identified in the affected muscle and subcutis. Small foci of
lymphocytic necrosis and generalized depletion were found
in the spleen in several frogs. A few perivascular clusters of
acidophils were noted in the meninges and rarely in the neuropil of the brain. Intestinal mucosa contained a mild lymphoplasmacytic infiltrate in a few cases.
With the exception of Streptococcus iniae, all of the bacterial species isolated have been reported previously as potential frog pathogens. Previously, S. iniae was reported as
an etiologic agent of disease in several fish species (Oreochromis spp. [tilapia],9 Lates calcarifer [barumunda],1 Morone chrysops ⫻ M. saxatilis [hybrid striped bass],9 Sciaenops ocellatus [red drum]2 and 2 mammals (humans11 and Inia
geoffrensis [freshwater dolphins].8 This is the first report of
S. iniae involvement in an amphibian disease outbreak.
Chryseobacterium meningosepticum was reported previously
as an etiologic agent in an epizootic from a colony of South
African clawed frogs (Xenopus laevis),5 and was isolated
from a colony of leopard frogs (Rana pipiens).10 This is the
first report of C. meningosepticum as a potential disease
agent in R. castesbeiana.
Granulomas noted in visceral organs were reminiscent of
mycobacteriosis, but acid-fast stains were constantly negative. Mycobacterium marinum has been reported previously
as a pathogen in frogs and other amphibians. Negative mycobacterial cultures also diminish the role of mycobacteria
in these cases. Acute bacterial sepsis or redleg in frogs has
historically been associated with A. hydrophila12; however,
several other species have been implicated in the pathogenesis of this syndrome.12 These have included Flavobacterium
(Chryseobacterium), Proteus, Pseudomonas, and other Aeromonas spp.4 Poor husbandry practices such as crowding,
poor water quality, spoiled food, and inappropriate cage design have been suggested as predisposing factors.12 Similar
clinical signs to those in this report, including anorexia, lethargy, cutaneous hyperemia, peripheral edema, subcutaneous
hemorrhage, cutaneous ulceration, and sudden death, have
been reported previously. Hepatic cord atrophy and hepatocellular vacuolar change can be associated with seasonality
and hibernation in frogs, but in these cases they likely resulted from anorexia. Neurologic clinical signs were considered to be secondary rather than primary, because only a
few frogs had histologic evidence of meningitis or encephalitis. Other clinical signs, including bloating or ascites, have
been reported, but were not detected in our cases. All of
these lesions are likely nonspecific findings of septicemia in
frogs and may indicate a preference for skin, subcutis, and
muscle to bacterial colonization.
Although antibiotic therapy instituted in individual aquaria successfully decreased mortalities in these outbreaks,
treatment of ponds was not undertaken because of cost and
logistical difficulties. Decreasing stress by limiting animal
densities and improving water quality may be the most effective preventative management. Traditionally, A. hydrophila has been considered the cause of bacterial sepsis or
redleg in frogs. However, the findings of this study suggest
that several opportunistic pathogens may be associated with
this clinical syndrome, emphasizing the importance of bac-
Brief Communications
terial cultures in identifying the particular bacterial species
responsible for a given epizootic.
6.
Sources and manufacturers
a. Trek Diagnostic Systems, Inc., Westlake, OH.
b. Remel, Norcross, GA.
c. API Analytab Products, Plainview, NY.
7.
8.
References
1. Bromage ES, Thomas A, Owens L: 1999, Streptococcus iniae,
a bacterial infection in barramundi Lates calcarifer. Dis Aquat
Org 36:177–181.
2. Eldar A, Perl S, Frelier PF, Bercovier H: 1999, Red drum
Sciaenops ocellatus mortalities associated with Streptococcus
iniae infections. Dis Aquat Org 36:121–127.
3. Gibbs EL, Gibbs TJ, Van Dyck PC: 1966, Rana pipiens: health
and disease. Lab Anim Care 16:142–160.
4. Glorioso JC, Amborski RL, Ambors GF: 1974, Microbiological
studies on septicemic bullfrogs (Rana catesbeiana). Am J Vet
Res 35:1241–1245.
5. Green SL, Bouley DM, Tolwani RJ, et al.: 1999, Identification
and management of an outbreak of Flavobacterium meningo-
9.
10.
11.
12.
433
septicum infection in a colony of South African clawed frogs
(Xenopus laevis). J Am Vet Med Assoc 214:1833–1838.
Miles EM: 1950, Red-leg in tree frogs caused by Bacterium
alkaligenes. J Gen Microbiol 4:434–436.
Olson ME, Gard S, Brown M, et al.: 1992, Flavobacterium indolgenes infection in leopard frogs. J Am Vet Med Assoc 201:
1766–1770.
Pier GB, Madin SH: 1976, Streptococcus iniae sp. nov., a betahemolytic Streptococcus isolated from an Amazon freshwater
dolphin, Inia geoffrensis. Int J Syst Bacterial 26:545–553.
Shoemaker CA, Klesius PH, Evans JJ: 2001, Prevalence of
Streptococcus iniae in tilapia, hybrid striped bass, and channel
catfish on commercial fish farms in the United States. Am J Vet
Res 62:174–177.
Taylor FR, Simmonds RC, Loeffler DG: 1993, Isolation of Flavobacterium meningosepticum in a colony of leopard frogs
(Rana pipiens). Lab Anim Sci 43:105.
Weinstein MR, Litt M, Kertesz DA, et al.: 1997, Invasive infections due to a fish pathogen Streptococcus iniae. N Engl J
Med 337:589–594.
Wright KM: 1996, Amphibian husbandry and medicine. In:
Reptile medicine and surgery, ed. Mader DR, pp. 448–449. WB
Saunders, Philadelphia, PA.
J Vet Diagn Invest 14:433–437 (2002)
A comparison of polymerase chain reaction with and without RNA extraction and
virus isolation for detection of bovine viral diarrhea virus in young calves
D. Deregt, P. S. Carman, R. M. Clark, K. M. Burton, W. O. Olson, S. A. Gilbert
Abstract. Previously, the authors described a multiplex reverse transcriptase–polymerase chain reaction
(PCR) assay for detection and typing of bovine viral diarrhea virus (BVDV) from blood of persistently infected
(PI) cattle that could be used with or without RNA extraction. In the present study, the PCR assay was evaluated
for its ability to detect BVDV in young calves as a screening tool for detection of persistent infections. Both
methods, PCR after RNA extraction (rPCR) and the direct method without RNA extraction (dPCR) were applied
and compared with virus isolation (VI) with diagnostic specimens. From 450 whole blood samples from Ontario
calves, 47 and 39 samples were positive by rPCR and VI, respectively. From the 47 samples positive by rPCR,
45 (96%) also were positive by dPCR when samples were tested both undiluted and diluted 1:10. In comparison
to VI, the relative sensitivities of both PCR assays were 100%. Examination of the results indicates that both
PCR assays can be used for screening calves for persistent infection with BVDV.
Bovine viral diarrhea virus (BVDV) causes a number of
diverse diseases in cattle including enteritis, reproductive
and respiratory disorders, hemorrhagic syndrome, persistent
infections, and mucosal disease.3 The importance of BVDV
as a pathogen is indicated by the large number of vaccines
(⬎140) that exist for the virus17 and the announcement in
recent years of BVDV eradication programs in some EuroFrom the Canadian Food Inspection Agency Lethbridge Laboratory, Animal Diseases Research Institute, PO Box 640, Lethbridge,
Alberta T1J 3Z4, Canada (Deregt, Clark, Burton, Olson, Gilbert),
and the Animal Health Laboratory, Laboratory Services Division,
University of Guelph, Box 3612, Guelph, Ontario N1H 6R8, Canada
(Carman).
Received for publication June 11, 2001.
pean countries.4 Whether controlled by vaccination or eradication (without vaccination), an important element in the
control of BVDV is the identification and removal of persistently infected (PI) animals from infected herds.
Bovine viral diarrhea virus is a member of the pestivirus
genus in the family Flaviviridae, a group of small enveloped
RNA viruses.21 Bovine viral diarrhea virus is now recognized as comprising 2 distinct genotypes or species, type 1
(BVDV I) and type 2 (BVDV II).15,16 Virulent strains of
BVDV II have been associated with severe disease outbreaks
of hemorrhagic syndrome and acute BVD in Canada and the
United States in recent years.9,15,16
Persistent infections may occur when the fetus becomes
infected during the first 4 months of gestation.2 Calves born
PI may appear normal or show stunted growth but invariably